Research article
Development of solid-phase extraction and methylation procedures to analyse free fatty acids in lipid-rich seeds

https://doi.org/10.1016/j.plaphy.2007.01.012Get rights and content

Abstract

In order to develop a sensitive and reliable method for FFA quantification in lipid matrices of seeds, two SPE procedures employed in meat and dairy chemistry were compared using a 100/1 mixture of triolein/heptadecanoic acid. The overall efficiency of the SPE procedure retained was satisfactory since it allowed removal of 99.8% of triacylglycerols (TAG) and recovery of 99.2% of FFA as quantified by gas chromatography of fatty acid methyl esters (FAME). However, the low amount of TAG eluted in the FFA fraction represented a non-negligible percentage (17%) of FAME and the procedure thus required further improvement. TAG pollution was successively decreased to 12%, 8% and finally 1.5% by: i) modifying the volume of elution of TAG; ii) removing the saponification step initially performed according to the standard FAME procedure; and iii) reducing the duration of the BF3-catalyzed methylation reaction to 1 min. The new SPE/methylation procedure described here was then compared to the most widely used method for FFA measurement in plants which is based on thin-layer chromatography (TLC). Both procedures were applied to coffee seeds stored for 0–18 months at 15 °C under 62% relative humidity and provided consistent results. A very clear negative correlation was observed between the loss of seed viability and the accumulation of FFA in seeds during the course of storage independent of the method employed for FFA quantification. However, we demonstrated that the TLC/on-silica methylation procedure underestimates FFA contents in comparison with the new SPE/methylation procedure because of a selective loss of unsaturated FA.

Introduction

De-esterification of fatty acids from glycerolipids, including phospholipids and triacylglycerols, has been identified as one of the major mechanisms of injury in various plant systems subjected to ageing, anoxia, drying, or freezing: e.g. potato cells under oxygen deprivation [16], dehydrated germinating soybean axes [24], rapidly dried carrot somatic embryos [27], aged Typha latifolia pollen [30], chilled non-orthodox neem seeds [22], and frozen Arabidopsis thaliana plants [33]. In particular, accumulation of free fatty acids (FFA) has been shown to play a crucial role in seed ageing [28], [31]. Two hypotheses have been proposed for the origin of FFA: free radical attack [25], [30] or lipase/phospholipase A2 activity [33]. FFA are well-established membrane destabilizing agents leading to irreversible damage [3], [12], [36] but are also known to play a crucial role in plant biotic and abiotic stress signalling [32].

In plant physiology studies, FFA are classically analyzed using the following procedure [16], [22], [24], [27], [28], [30]: FFA are separated from other lipid classes by thin-layer chromatography (TLC), scraped off TLC plates, derivatized (generally by methylation) and quantified by gas chromatography (GC) using heptadecanoic acid (17:0) as internal standard. This method is called thereafter the “TLC/on-silica methylation” procedure. However, Sowa and Subbaiah [26] recently demonstrated that some widespread brands of TLC plates lead to selective loss of unsaturated FA when FA are methylated in presence of the silica gel scraped from the plate, as is usually the case in plant physiology studies. These authors suggested that loss of unsaturated FA could be circumvented by adding a supplementary liquid-liquid extraction step after TLC separation [26].

A useful alternative method consists in the selective derivatization of FFA without their prior purification using diazomethane, which is not effective for trans-methylation of esterified FA [14]. However, because diazomethane is potentially explosive, is only stable for very short periods and needs to be prepared on-site with carcinogenous products [2], this method has only very rarely been used in the plant physiology community [35].

Very recently, a new method using electrospray ionization mass spectrometry (ESI-MS/MS) was successfully applied to lipid species profiling in plants [8], [34]. This high-throughput approach requires only simple preparation and small samples to identify and quantify a wide range of lipid molecular species. It allows FFA quantification without their prior separation and derivatization. However, the major drawback of the ESI-MS/MS technique is the considerable cost of the equipment in comparison with GC devices.

Finally, another strategy based on the use of solid-phase extraction (SPE) was recently developed in food chemistry for FFA purification in dairy and meat products [19], [21], [29]. With the arrival of disposable cartridges (e.g. silica, aminopropyl, cyanopropyl, diol, C8, C18), SPE is now preferred to TLC because it is simpler, faster, and more reproducible, and allows fractionation of higher amounts of lipids [20]. Though SPE methods dedicated to plant physiology research have already been used, for example to separate plant membrane lipids [17], to our knowledge, the use of SPE for FFA analysis in plant tissues has never been reported.

The objective of the present work was to develop a SPE-GC method to isolate and measure FFA in lipid-rich seeds for immediate use in our investigations on the non-orthodox storage behaviour of numerous tropical seeds, such as Citrus [10] and coffee [4] seeds, which generally contain large amounts of lipids. A comprehensive survey of the literature on the use of SPE for lipid fractionation allowed us to identify two reference methods developed by Kaluzny et al. [11], and Prieto et al. [15]. These two distinct procedures, based on the use of aminopropyl and silica stationary phases, respectively, have both been widely used in the food chemistry community (e.g. refs. [1], [9], [19], [21], [23], [29]). In the present study, we first tested the applicability of these two reference methods [11], [15] to FFA isolation in seed lipid matrices. The SPE method providing the best results, together with the FA methylation procedure were then further refined for accurate measurement of FFA in lipid-rich seeds. The method was finally compared to the TLC/on-silica methylation procedure [16], [22], [24], [27], [28], [30] and validated in ageing coffee seeds.

Section snippets

Results and discussion

The two lipid fractionation SPE methods compared in the present work were originally developed by Kaluzny et al. [11] and Prieto et al. [15], respectively, and are presented in Table 1. Although the two methods differ with respect to the stationary phase and the elution solvents, they include the same elution sequence: neutral lipids in Fraction 1 (mainly mono-, di- and tri-acylglycerols and sterol esters in seeds), FFA in Fraction 2, and finally polar lipids in Fraction 3. In oily seeds, the

Conclusion

The present work describes the step-by-step optimization of a new procedure combining SPE and GC for FFA analysis in lipid-rich seeds. After a comparison of two general SPE methods, modification of the elution volume of neutral lipids resulted in a decrease in TAG pollution from about 17 to 12%. Removing the saponification step of the standard procedure ISO-5509 for FAME preparation led to a further decrease in the percentage of FAME deriving from TAG from about 8% to 4%. Finally, percentage

Plant material

Fresh mature seeds of Coffea arabica (variety Caturra) were obtained from CICAFE, San Jose, Costa Rica. The initial water content of the seeds was 0.58 g H2O g−1 dw. Seeds were desiccated at 15 °C in the dark under 62% RH and stored for 18 months under the same conditions. Seed viability was assessed after 0, 6, 12 and 18 months of storage using the criteria of normal seedling development, i.e. emergence of the hypocotyl, radicle geotropic growth and opening of cotyledonary leaves, after 6 weeks of

Acknowledgements

The authors acknowledge the Commission of the European Union for funding support through CRYMCEPT (“Establishing Cryopreservation Methods for Conserving European Plant Germplasm Collections”, Quality of Life and Management of Living Resources, QLK5-CT-2002-01279).

References (36)

Cited by (0)

View full text